SIGCOMM '93 Conference proceedings on Communications architectures, protocols and applications
Performance analysis of reconstruction algorithms for packet voice communications
Computer Networks and ISDN Systems
A Dynamic Priority Assignment Technique for Streams with (m, k)-Firm Deadlines
IEEE Transactions on Computers
Real-Time Communication in Multihop Networks
IEEE Transactions on Parallel and Distributed Systems
LCN '97 Proceedings of the 22nd Annual IEEE Conference on Local Computer Networks
Graceful Degradation in Real-time Control Applications Using (m, k)-Firm Guarantee
FTCS '97 Proceedings of the 27th International Symposium on Fault-Tolerant Computing (FTCS '97)
A scheme for real-time channel establishment in wide-area networks
IEEE Journal on Selected Areas in Communications
Scalability and QoS guarantee for streams with (m,k)-firm deadline
Computer Standards & Interfaces
Hi-index | 0.24 |
In this paper, we address the problem of best-effort scheduling of (m,k)-firm real-time streams in multihop networks. The existing solutions for the problem ignore scalability considerations because the solutions maintain a separate queue for each stream and maintain per-stream state information. In this context, we propose a scheduling algorithm, Enhanced Distance Based Priority (EDBP), which is scalable (fixed scheduling cost) while offering performance close to that of the existing solutions. The proposed EDBP algorithm achieves this by allowing multiplexing of streams into a fixed number of queues and by using the notion of a look-ahead window. In the EDBP algorithm, at any point of time, the best packet for transmission is selected based on the state of the stream combined with the laxity of the packet. We have conducted extensive simulation studies to evaluate the effectiveness of the proposed EDBP algorithm by comparing it with existing algorithms for the problem. The simulation studies reveal that the dynamic failure performance of the EDBP algorithm is very close to that of the existing algorithms with a significant reduction in scheduling cost.